an in depth look at gold

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

an in depth look at gold

Post by larryswedroe »

http://www.etf.com/sections/index-inves ... ds-glitterd

What exactly is its role? Does it hedge inflation (yes and no), and what's its "value"?

Hope you find it helpful
Larry
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: an in depth look at gold

Post by nedsaid »

It is hard to know what the intrinsic value of gold is. Rick Ferri's guess of $600 to $800 an ounce or the $836 cited in the article is as good as any. It is an entirely subjective judgment as gold generates no earnings or cash flow. I am not buying though I have a twinge of regret for not purchasing a gold or precious metals mutual fund a couple years ago. The U.S. Stock market has been doing well, so I haven't missed out. Not only that, but the long term investment record of gold and precious metals and the funds of their mining stocks is not very good. I need better than a zero real rate of return over my lifetime.
A fool and his money are good for business.
User avatar
pennstater2005
Posts: 2509
Joined: Wed Apr 11, 2012 8:50 pm

Re: an in depth look at gold

Post by pennstater2005 »

Gold does well, polls show favorable as an investment. Gold does bad, polls show less than favorable as an investment. Same with stocks and same with bonds I imagine. Emotional investing and investors...no surprise there.
Last edited by pennstater2005 on Fri Mar 17, 2017 10:19 pm, edited 1 time in total.
“If you think nobody cares if you're alive, try missing a couple of car payments.” – Earl Wilson
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: an in depth look at gold

Post by willthrill81 »

While it's real value over long periods of time isn't likely to go up, it still has had investment value when it comes to smoothing out returns sufficiently to increase SWRs beyond stock and bond only portfolios, at least for the last ~40 years.
The Sensible Steward
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: an in depth look at gold

Post by nedsaid »

Money under my mattress does not correlate with stocks and yet I feel no need to buy a "money under the mattress" mutual fund. If I buy a non-correlating asset class to stocks, I would at least like to have some return from it. Zero percent real return over long periods of time does not sound appealing.
A fool and his money are good for business.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: an in depth look at gold

Post by willthrill81 »

nedsaid wrote:Money under my mattress does not correlate with stocks and yet I feel no need to buy a "money under the mattress" mutual fund. If I buy a non-correlating asset class to stocks, I would at least like to have some return from it. Zero percent real return over long periods of time does not sound appealing.
It may not 'sound' appealing, but I encourage you to look at what the impact of a relatively small position in gold (~10%) would have done to a portfolio's maximum SWR over the last 40 years. History may not repeat itself, but we're all banking on it doing so to some extent.
The Sensible Steward
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: an in depth look at gold

Post by lack_ey »

willthrill81 wrote:While it's real value over long periods of time isn't likely to go up, it still has had investment value when it comes to smoothing out returns sufficiently to increase SWRs beyond stock and bond only portfolios, at least for the last ~40 years.
A period of positive real returns for gold over which the pullbacks (1980-2001, 2012-present) coincided with huge stock bull markets. How meaningful and representative of the future do you expect that observed relationship to be going forward?
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: an in depth look at gold

Post by nedsaid »

willthrill81 wrote:
nedsaid wrote:Money under my mattress does not correlate with stocks and yet I feel no need to buy a "money under the mattress" mutual fund. If I buy a non-correlating asset class to stocks, I would at least like to have some return from it. Zero percent real return over long periods of time does not sound appealing.
It may not 'sound' appealing, but I encourage you to look at what the impact of a relatively small position in gold (~10%) would have done to a portfolio's maximum SWR over the last 40 years. History may not repeat itself, but we're all banking on it doing so to some extent.
The Harry Brown Permanent Portfolio is 25% stocks, 25% cash, 25% long-term treasuries, and 25% gold. My understanding is that this portfolio has done well over time. There is something to your point. I just have never pulled the trigger and bought gold for my portfolio. If it was under $600 an ounce, I might think about it.
A fool and his money are good for business.
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: an in depth look at gold

Post by lack_ey »

Wait a sec, I just skimmed the Erb and Harvey update paper.

Did they seriously just look at subsequent return of gold over their entire dataset using real price (relative to historical average computed over that same dataset) as an explanatory variable?

That has so many methodological problems I don't even feel like addressing them in any detail so I'll take the lazy way out and link philsophicaleconomics:
http://www.philosophicaleconomics.com/2 ... e-fitting/

I don't really see any real support behind the idea of the golden constant, either. Is the real average price over their data sample that much better a signal for what baseline gold price should look like than the actual current price?
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: an in depth look at gold

Post by larryswedroe »

If gold is an inflation hedge then current price relative to inflation index should be a good indicator of it's value, again assuming that's it's purpose. That seems like a simple and logic concept since it is in fact why many buy it.
Larry
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: an in depth look at gold

Post by willthrill81 »

nedsaid wrote:
willthrill81 wrote:
nedsaid wrote:Money under my mattress does not correlate with stocks and yet I feel no need to buy a "money under the mattress" mutual fund. If I buy a non-correlating asset class to stocks, I would at least like to have some return from it. Zero percent real return over long periods of time does not sound appealing.
It may not 'sound' appealing, but I encourage you to look at what the impact of a relatively small position in gold (~10%) would have done to a portfolio's maximum SWR over the last 40 years. History may not repeat itself, but we're all banking on it doing so to some extent.
The Harry Brown Permanent Portfolio is 25% stocks, 25% cash, 25% long-term treasuries, and 25% gold. My understanding is that this portfolio has done well over time. There is something to your point. I just have never pulled the trigger and bought gold for my portfolio. If it was under $600 an ounce, I might think about it.
While I'm not a big fan of the PP for reasons beyond its gold holding, its real returns for the last 40 years have been around 4.7% annually with a std. dev. of 6.8%. That's well behind the TSM over the same period (7.4% for it), but the PP also had a std. dev. of 6.8% instead of 17.2%. But even with that lower return, the PP would have supported a SWR around 1% higher than a 100% TSM portfolio.

I think that the Golden Butterfly is a better portfolio all-around. Its return over the same period was 6.1% with a std. dev. of 7.6%.

I wouldn't implement either of these portfolios until near the end of the accumulation phase, but during the withdrawal phase, even 'marginal' returns such as that of the PP are more than offset in terms of the SWR by their low volatility. Most people far underestimate the negative impact of volatility on SWRs; it's at least as impactful as returns.

And for those who are just dead set against gold, it could be substituted with commodities in the above portfolios and still have achieved remarkably similar performance in terms of volatility reduction.
The Sensible Steward
lazyday
Posts: 3849
Joined: Wed Mar 14, 2007 10:27 pm

Re: an in depth look at gold

Post by lazyday »

willthrill81 wrote:While it's real value over long periods of time isn't likely to go up, it still has had investment value when it comes to smoothing out returns sufficiently to increase SWRs beyond stock and bond only portfolios, at least for the last ~40 years.
Are there any SWR studies that have included PM equity or gold miners as an asset class?
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: an in depth look at gold

Post by willthrill81 »

lazyday wrote:
willthrill81 wrote:While it's real value over long periods of time isn't likely to go up, it still has had investment value when it comes to smoothing out returns sufficiently to increase SWRs beyond stock and bond only portfolios, at least for the last ~40 years.
Are there any SWR studies that have included PM equity or gold miners as an asset class?
I'm not aware of any academic research, but there are a few articles at Portfolio Charts where it is discussed and the data examined.

https://portfoliocharts.com/2015/09/08/ ... bly-wrong/

This is one of those areas where I think that there is more potential upside than downside. There isn't much to lose by moving say 10% of a retirement portfolio into gold, but there seems to be strong potential for a lot of upside in the form of reduced overall volatility. And as I've pointed out already, commodities would have achieved largely the same effect over the last 40+ years.
The Sensible Steward
IngognitoUSA
Posts: 206
Joined: Sat Mar 11, 2017 6:54 am

Re: an in depth look at gold

Post by IngognitoUSA »

Physical gold as investment in not very good. However, if you had to flee Syria the only thing worthwhile you can carry out would be gold. Your property, stock investment and currency are probably worthless.

If your currency got demonetized as in India recently, gold would have been a good alternative to keep.
lazyday
Posts: 3849
Joined: Wed Mar 14, 2007 10:27 pm

Re: an in depth look at gold

Post by lazyday »

willthrill81 wrote:I'm not aware of any academic research, but there are a few articles at Portfolio Charts where it is discussed and the data examined.

https://portfoliocharts.com/2015/09/08/ ... bly-wrong/
Nice it lets you calculate SWR, though gold here is the metal not equity which I prefer.
There isn't much to lose by moving say 10% of a retirement portfolio into gold, but there seems to be strong potential for a lot of upside in the form of reduced overall volatility.
I hate gold but as a retiree I do own a couple percent of gold miners for essentially this reason. Bill Bernstein has written about even a small amount being worthwhile because of volatility. Maybe in a day of deep pessimism and low prices I’ll increase to 5%.
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: an in depth look at gold

Post by packer16 »

I think when you look at gold you need to remove the initial period of 1976 to 1979/1980 as gold was set free from the value of the USD at the beginning of the period so you need to account for the transition from a fixed price to a floating price. If you want to see gold long-term returns versus other asset see Fig 12.1 in The Future for Investors by Siegel which shows over 200 years that gold returned .2% per year versus 2.9% per year fro t-bills. With inflation at 1.3% inflation-adjusted gold has actual declined in real terms by 1.2% versus 1.6% gain for t-bills. So over the long-term with gold you a fighting a good sized head wind much like when you are buying a fund of shorted stocks. The main issue I see with using historical data when looking at correlation is you need to have the same conditions repeat.

Second, long-term gold and and most commodities (except O&G) go down in value in real sense as the supply ever increasing. Oil & gas are consumed so it actually disappears. As far as I can see, you can use t-bills or high grade corporate paper to act as your inflation hedge as you will actual get inflation plus a small premium that will not decline when commodities continue there inevitable decline. So IMO you get the same return with less volatility. The key in my mind you are buying either a productive asset (stocks, real estate or bonds/bills) or a non-productive one, commodities.

Packer
Buy cheap and something good might happen
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: an in depth look at gold

Post by willthrill81 »

packer16 wrote:I think when you look at gold you need to remove the initial period of 1976 to 1979/1980 as gold was set free from the value of the USD at the beginning of the period so you need to account for the transition from a fixed price to a floating price. If you want to see gold long-term returns versus other asset see Fig 12.1 in The Future for Investors by Siegel which shows over 200 years that gold returned .2% per year versus 2.9% per year fro t-bills. With inflation at 1.3% inflation-adjusted gold has actual declined in real terms by 1.2% versus 1.6% gain for t-bills. So over the long-term with gold you a fighting a good sized head wind much like when you are buying a fund of shorted stocks. The main issue I see with using historical data when looking at correlation is you need to have the same conditions repeat.

Second, long-term gold and and most commodities (except O&G) go down in value in real sense as the supply ever increasing. Oil & gas are consumed so it actually disappears. As far as I can see, you can use t-bills or high grade corporate paper to act as your inflation hedge as you will actual get inflation plus a small premium that will not decline when commodities continue there inevitable decline. So IMO you get the same return with less volatility. The key in my mind you are buying either a productive asset (stocks, real estate or bonds/bills) or a non-productive one, commodities.

Packer
While the very long-term returns of gold and TIPS, for instance, may be essentially the same, that does not mean that they will have similar performance in terms of volatility and price movement in the short-term.

People think that because gold has virtually no long-term real return that it has no value, but that's not the only measure of performance, especially when you're in the withdrawal phase. Volatility absolutely kills your safe withdrawal rate. Even though it's not my favorite portfolio, the PP would have achieved a 1% higher SWR (~25% higher) over the last 40+ years than a 60% TSM / 40% TBM portfolio because it's volatility is far less, and it achieved that with only 25% in stocks. Some state that this might only be an anomaly, which is true, but it's hard to say that 40 years of data are a cumulative anomaly.

Whether it's gold, commodities, or something else, reducing volatility in the withdrawal phase has a substantial positive effect on your SWR, even if it comes partly at the expense of returns.
The Sensible Steward
lazyday
Posts: 3849
Joined: Wed Mar 14, 2007 10:27 pm

Re: an in depth look at gold

Post by lazyday »

Just to be clear, gold equity has high volatility, but with its low correlation to other portfolio assets you might use it to construct a portfolio of (hopefully) lower volatility or higher returns.

The high volatility means that even a little can make a difference, and is theoretically a benefit because of the low correlation.

I'm less familiar with the arguments for physical gold.
Last edited by lazyday on Sun Mar 19, 2017 7:57 am, edited 1 time in total.
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: an in depth look at gold

Post by packer16 »

Withdraw needs to be examined in context of how you are going to withdraw & how flexible your spending is. In the commentary that I have heard from many folks here in retirement is that spending flexibility can be as or more important than withdraw strategies. This is probably analogous to spending flexibility when you are saving.

Second there are withdraw strategies to deal with volatility, like withdrawing from a risky asset pool when it increases (most years) & have a non-risk pool to withdraw from when the risky asset declines. The assumption of volatility being a concern is based upon a fixed spending requirement & an withdraw that pulls equally from the portfolio across assets classes. I think both of these can be mitigated for the most part by techniques described above.

In essence these low expected return assets are hedges. Hedges can be expensive if they are held for long periods of time. If you can identify when they are most needed (probably right before retirement & for a few years after), then you can buy for that window similar to insurance where you only buy it when you need it.

Then the question is what assets to invest in based upon expected return primarily and gold & commodities fall out pretty quick.
Buy cheap and something good might happen
User avatar
TheTimeLord
Posts: 12130
Joined: Fri Jul 26, 2013 2:05 pm

Re: an in depth look at gold

Post by TheTimeLord »

larryswedroe wrote:http://www.etf.com/sections/index-inves ... ds-glitterd

What exactly is its role? Does it hedge inflation (yes and no), and what's its "value"?

Hope you find it helpful
Larry
My impression has been gold is a currency and it is only an inflation hedge if that inflation is caused by the weakening of the currency it is being priced in. So I view it primary value is the protection against a deeper than normal financial crisis.
IMHO, Investing should be about living the life you want, not avoiding the life you fear. | Run, You Clever Boy! [9085]
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: an in depth look at gold

Post by NiceUnparticularMan »

Citing 40 or 40+ years specifically is a curious thing to do, and as it happens that is what is required to take the history back to the exact period where the Bretton Woods system was unwinding, allowing the rapid appreciation of gold in USD terms. In a few years from now, suddenly the trailing 40 years will look much worse for gold, but then presumably we will be hearing about the last 45 years instead of 40, and so on.

Of course in general it is good to take an open-minded view of what is possible in the future. But the unwinding of the Bretton Woods system was a very specific sort of historic event that required certain prior conditions that do not exist today. Therefore in this particular case it is probably unwise to implicitly assume that same sort of event could occur again any moment.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: an in depth look at gold

Post by NiceUnparticularMan »

TheTimeLord wrote:My impression has been gold is a currency and it is only an inflation hedge if that inflation is caused by the weakening of the currency it is being priced in. So I view it primary value is the protection against a deeper than normal financial crisis.
My thought is if idiosyncratic USD devaluation is what you are concerned about, you could simply hold a basket of non-USD-denominated assets, maybe the equivalent of T-bills. But it doesn't seem like you need a specific allocation for that purpose if you are holding a diverse set of unhedged non-USD-denominated assets already.

Personally, it makes me nervous that the monetary value of gold is essentially a matter of convention. If the world suddenly or gradually decides to unwind that convention and gold reverts to its industrial/aesthetic value only, it could perform quite poorly.
lazyday
Posts: 3849
Joined: Wed Mar 14, 2007 10:27 pm

Re: an in depth look at gold

Post by lazyday »

NiceUnparticularMan wrote:Personally, it makes me nervous that the monetary value of gold is essentially a matter of convention. If the world suddenly or gradually decides to unwind that convention and gold reverts to its industrial/aesthetic value only, it could perform quite poorly.
This is the primary reason I hate gold. To me it's a pure Greater Fool asset.

Yes, it has been for millennia, but that can change with technology. Such as bitcoin or __coin becoming a better store of value, or the price becoming too unreliable (or low!) with massive supplies from mining under the ocean.
User avatar
TheTimeLord
Posts: 12130
Joined: Fri Jul 26, 2013 2:05 pm

Re: an in depth look at gold

Post by TheTimeLord »

NiceUnparticularMan wrote:
TheTimeLord wrote:My impression has been gold is a currency and it is only an inflation hedge if that inflation is caused by the weakening of the currency it is being priced in. So I view it primary value is the protection against a deeper than normal financial crisis.
My thought is if idiosyncratic USD devaluation is what you are concerned about, you could simply hold a basket of non-USD-denominated assets, maybe the equivalent of T-bills. But it doesn't seem like you need a specific allocation for that purpose if you are holding a diverse set of unhedged non-USD-denominated assets already.

Personally, it makes me nervous that the monetary value of gold is essentially a matter of convention. If the world suddenly or gradually decides to unwind that convention and gold reverts to its industrial/aesthetic value only, it could perform quite poorly.
This is true but you would have gold bugs argue the same about fiat currencies. It seems once you go beyond a barter system to using a currency you run that risk. Value is always subjective and often situational. I am arguing neither for or against holding gold just stating my belief that what gold is at this point is an alternate currency that can't just be randomly created by a governmental agency. If holding such an asset makes sense is entirely an individual decision.
IMHO, Investing should be about living the life you want, not avoiding the life you fear. | Run, You Clever Boy! [9085]
User avatar
TheTimeLord
Posts: 12130
Joined: Fri Jul 26, 2013 2:05 pm

Re: an in depth look at gold

Post by TheTimeLord »

lazyday wrote:
NiceUnparticularMan wrote:Personally, it makes me nervous that the monetary value of gold is essentially a matter of convention. If the world suddenly or gradually decides to unwind that convention and gold reverts to its industrial/aesthetic value only, it could perform quite poorly.
This is the primary reason I hate gold. To me it's a pure Greater Fool asset.

Yes, it has been for millennia, but that can change with technology. Such as bitcoin or __coin becoming a better store of value, or the price becoming too unreliable (or low!) with massive supplies from mining under the ocean.
Last I heard that existing mined amount of gold in the world still amounted to less than 1 ounce per person on the planet so if it is indeed a currency I would think making it more readily available might actually make it more desirable. Now if it is commodity I would think the opposite would be true.
IMHO, Investing should be about living the life you want, not avoiding the life you fear. | Run, You Clever Boy! [9085]
lazyday
Posts: 3849
Joined: Wed Mar 14, 2007 10:27 pm

Re: an in depth look at gold

Post by lazyday »

TheTimeLord wrote: so if it is indeed a currency I would think making it more readily available might actually make it more desirable
Interesting argument, but I'm not sure that gold is quite a currency? I haven't read much about gold, but have the idea that many people use it to store wealth, as a hedge/diversifier, or for speculation, but generally not as cash for smaller transactions. So I don't think more of it would help a lot, if you want to store $100 as gold today, you just get a tiny piece.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: an in depth look at gold

Post by NiceUnparticularMan »

TheTimeLord wrote:This is true but you would have gold bugs argue the same about fiat currencies. It seems once you go beyond a barter system to using a currency you run that risk. Value is always subjective and often situational. I am arguing neither for or against holding gold just stating my belief that what gold is at this point is an alternate currency that can't just be randomly created by a governmental agency. If holding such an asset makes sense is entirely an individual decision.
Oh, there are certainly similar issues regarding fiat currencies. That is among the reasons my long-term investment plan does not include bricks of paper currency, and generally when formulating that plan I thought a bit about various inflation/devaluation scenarios and how to provide against them.

I am not going to try to persuade gold fans to change their ways, but from my perspective this sort of argument for holding gold is philosophically equivalent to something like arguing that I should be holding a portion of my portfolio in bricks of Swiss francs. It is going to take quite a bit of convincing for me to believe that is a sensible way of dealing with these issues, and it won't be persuasive just because someone has done some back-testing over a non-random period of years unique to Swiss history that shows such a holding would have helped improve returns.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: an in depth look at gold

Post by NiceUnparticularMan »

lazyday wrote:Interesting argument, but I'm not sure that gold is quite a currency? I haven't read much about gold, but have the idea that many people use it to store wealth, as a hedge/diversifier, or for speculation, but generally not as cash for smaller transactions. So I don't think more of it would help a lot, if you want to store $100 as gold today, you just get a tiny piece.
Yeah, the historic reasons why gold was actually practically useful as a means of exchange seem rather diminished these days.
Commodore
Posts: 18
Joined: Sun Aug 30, 2015 5:39 pm

Re: an in depth look at gold

Post by Commodore »

One thing that prevents me from investing in gold is technology. Being in the commodities business, we have seen how massive technological breakthroughs when extracting fossil fuels (Oil & Gas) has caused the overall pricing complex to be completely re-set to lower production costs and ultimately lower commodity prices. If this happened in gold mining, gold prices would in turn re-set to much lower levels. This is a huge downside risk IMO.
JFP_SF
Posts: 76
Joined: Sat Oct 01, 2016 12:01 pm

Re: an in depth look at gold

Post by JFP_SF »

NiceUnparticularMan wrote:
I am not going to try to persuade gold fans to change their ways, but from my perspective this sort of argument for holding gold is philosophically equivalent to something like arguing that I should be holding a portion of my portfolio in bricks of Swiss francs. It is going to take quite a bit of convincing for me to believe that is a sensible way of dealing with these issues, and it won't be persuasive just because someone has done some back-testing over a non-random period of years unique to Swiss history that shows such a holding would have helped improve returns.
All arguments based on back-testing use a non-random period of years unique to some history. So, that's not much of an argument against anything.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: an in depth look at gold

Post by NiceUnparticularMan »

JFP_SF wrote:All arguments based on back-testing use a non-random period of years unique to some history. So, that's not much of an argument against anything.
Well, certainly many arguments based on back-testing have some general form of this problem. And generally, I am not a huge fan of relying on back-testing because I think economies and financial systems are subject to evolution, including structural breaks. I think it can be useful to the extent it can illustrate how things can go wrong, but I believe for optimization purposes it is not really all that powerful of a tool.

That said, you can combat this issue with various social science techniques, like dividing your data into various subsets, then building your optimization model based on the data in one subset, then doing an out of sample test to see if the model holds up. This takes a lot of work and self-discipline, and at the end of the day still might not be all that illuminating or reliable.

However, in this specific case I was referring to choosing a period like 40 years for back-testing the benefits of holding gold, when we know there was a unique event specific to gold happening around 40 years ago that made holding gold favorable in that period (at least if you are looking at returns in USD). So even if it is hard to design a really bulletproof back-testing protocol, we know for a fact that particular one is really not going to be reliable for this particular purpose.

Although I did not flesh out the analogy, what I have in mind is something like knowing that X number of years ago, something unique and non-repeatable happened to make holding Swiss francs advantageous for USD-focused investors. What you would definitely not want to do is then back-test portfolios including Swiss francs that specifically looked at the period starting X years ago.
JFP_SF
Posts: 76
Joined: Sat Oct 01, 2016 12:01 pm

Re: an in depth look at gold

Post by JFP_SF »

NiceUnparticularMan wrote:
JFP_SF wrote:All arguments based on back-testing use a non-random period of years unique to some history. So, that's not much of an argument against anything.
Well, certainly many arguments based on back-testing have some general form of this problem. And generally, I am not a huge fan of relying on back-testing because I think economies and financial systems are subject to evolution, including structural breaks. I think it can be useful to the extent it can illustrate how things can go wrong, but I believe for optimization purposes it is not really all that powerful of a tool.

That said, you can combat this issue with various social science techniques, like dividing your data into various subsets, then building your optimization model based on the data in one subset, then doing an out of sample test to see if the model holds up. This takes a lot of work and self-discipline, and at the end of the day still might not be all that illuminating or reliable.

However, in this specific case I was referring to choosing a period like 40 years for back-testing the benefits of holding gold, when we know there was a unique event specific to gold happening around 40 years ago that made holding gold favorable in that period (at least if you are looking at returns in USD). So even if it is hard to design a really bulletproof back-testing protocol, we know for a fact that particular one is really not going to be reliable for this particular purpose.

Although I did not flesh out the analogy, what I have in mind is something like knowing that X number of years ago, something unique and non-repeatable happened to make holding Swiss francs advantageous for USD-focused investors. What you would definitely not want to do is then back-test portfolios including Swiss francs that specifically looked at the period starting X years ago.
I still think the argument fails. All back-tested periods have unique unrepeatable events that affected assets. Some of them are major like the Russian Revolution's effect on Russian stocks and bonds and some of them are more minor like the advent of Index funds. Prior to the first index fund, holding the S&P 500 was impossible, but all the back-testing ignores this. I could just as well argue that the we should ignore the returns in stocks since 2008, because of the unprecedented and unrepeatable monetary policy that has occurred since then, if I accept your arguments on gold in the 70s.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: an in depth look at gold

Post by NiceUnparticularMan »

JFP_SF wrote:I still think the argument fails. All back-tested periods have unique unrepeatable events that affected assets. Some of them are major like the Russian Revolution's effect on Russian stocks and bonds and some of them are more minor like the advent of Index funds. Prior to the first index fund, holding the S&P 500 was impossible, but all the back-testing ignores this. I could just as well argue that the we should ignore the returns in stocks since 2008, because of the unprecedented and unrepeatable monetary policy that has occurred since then, if I accept your arguments on gold in the 70s.
From my perspective, you are not showing the argument is failing with respect to back-testing gold for the specific period of the last 40 years. You are showing instead that we should be very cautious about back-testing in general, a point with which I agree.

Just so everyone knows what we are talking about--under the Bretton Woods international monetary system negotiated after WWII, the price of gold in USD was pegged at $35. By the 1960s, the USD had become very important internationally, and that peg had gotten to be much lower than where the free-floating price would be, and this was contributing to various international monetary problems. So, we transitioned to a system in which the price of gold was free-floating in USD, which continues today. Due to the complexities of the transition, this took several years.

40 years ago, the transition had started but was not complete. Predictably, this meant that gold was going to appreciate in USD, because that $35 peg had been way below the free-floating rate.

Unless and until we go back to pegging the price of gold in USD, this isn't going to happen again. From my perspective, that makes data from this period highly suspect, and it is particularly suspect to actually begin our back-testing right when this is happening.

Again, I realize the same sort of argument could be made about many suggested back-tests. I don't see that as a flaw with the argument, I see that as a reason to be cautious about back-testing in general.
JFP_SF
Posts: 76
Joined: Sat Oct 01, 2016 12:01 pm

Re: an in depth look at gold

Post by JFP_SF »

NiceUnparticularMan wrote:
JFP_SF wrote:I still think the argument fails. All back-tested periods have unique unrepeatable events that affected assets. Some of them are major like the Russian Revolution's effect on Russian stocks and bonds and some of them are more minor like the advent of Index funds. Prior to the first index fund, holding the S&P 500 was impossible, but all the back-testing ignores this. I could just as well argue that the we should ignore the returns in stocks since 2008, because of the unprecedented and unrepeatable monetary policy that has occurred since then, if I accept your arguments on gold in the 70s.
From my perspective, you are not showing the argument is failing with respect to back-testing gold for the specific period of the last 40 years. You are showing instead that we should be very cautious about back-testing in general, a point with which I agree.

Just so everyone knows what we are talking about--under the Bretton Woods international monetary system negotiated after WWII, the price of gold in USD was pegged at $35. By the 1960s, the USD had become very important internationally, and that peg had gotten to be much lower than where the free-floating price would be, and this was contributing to various international monetary problems. So, we transitioned to a system in which the price of gold was free-floating in USD, which continues today. Due to the complexities of the transition, this took several years.

40 years ago, the transition had started but was not complete. Predictably, this meant that gold was going to appreciate in USD, because that $35 peg had been way below the free-floating rate.

Unless and until we go back to pegging the price of gold in USD, this isn't going to happen again. From my perspective, that makes data from this period highly suspect, and it is particularly suspect to actually begin our back-testing right when this is happening.

Again, I realize the same sort of argument could be made about many suggested back-tests. I don't see that as a flaw with the argument, I see that as a reason to be cautious about back-testing in general.
We definitely agree about the need to be cautious in trusting back-testing.

We'll have to agree to disagree about the validity of ignoring the 70s as applied to gold's return in dollars. Overall, the period since the 70s has been an unprecedented era of fiat money in general. That's had distorting effects on every asset.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: an in depth look at gold

Post by NiceUnparticularMan »

JFP_SF wrote:We definitely agree about the need to be cautious in trusting back-testing.

We'll have to agree to disagree about the validity of ignoring the 70s as applied to gold's return in dollars. Overall, the period since the 70s has been an unprecedented era of fiat money in general. That's had distorting effects on every asset.
Well, we could compromise on ignoring all of it!

As I suggested previously, from my perspective there is no prima facie appeal to holding non-productive assets like bricks of paper money or chunks of gold in a long-term investment portfolio. If the argument for doing so depends on an unreliable back-test, then I am just going to stick with that as my answer.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: an in depth look at gold

Post by willthrill81 »

NiceUnparticularMan wrote:As I suggested previously, from my perspective there is no prima facie appeal to holding non-productive assets like bricks of paper money or chunks of gold in a long-term investment portfolio.
I think you've hit the nail on the head as to why so many Bogleheads are opposed to gold. It has everything to do with the fact that it's an inanimate object. I don't think that there's a logical reason for this disdain, however, if it can be shown that gold will (or would have) improve your investment performance. That leads into your second comment.
NiceUnparticularMan wrote:If the argument for doing so depends on an unreliable back-test, then I am just going to stick with that as my answer.
Like everyone else, I have no idea what the price of gold will be in the next forty years, but I do know what it has done for the last 40 years. Unless we were restricting our analysis to those countries whose currency was not connected to gold, I find any analysis of gold prices prior to that highly unreliable. But 40 years of data is nothing to sneeze at; that's longer than most of us will be in the accumulation phase at the least. And those data are quite clear; during the last 40 years, a small holding of gold (up to ~20% it seems to me) in an otherwise 'typical' portfolio with stocks and bonds would have reduced the portfolio's volatility without a corresponding reduction in returns such that it the safe withdrawal rate would be significantly higher than a portfolio comprised of only stocks and bonds in any ratio.

Now one could argue about whether that will continue to be the case going forward, but there's no disputing that it was true for the last 40 years.
The Sensible Steward
JFP_SF
Posts: 76
Joined: Sat Oct 01, 2016 12:01 pm

Re: an in depth look at gold

Post by JFP_SF »

willthrill81 wrote:
NiceUnparticularMan wrote:As I suggested previously, from my perspective there is no prima facie appeal to holding non-productive assets like bricks of paper money or chunks of gold in a long-term investment portfolio.
I think you've hit the nail on the head as to why so many Bogleheads are opposed to gold. It has everything to do with the fact that it's an inanimate object. I don't think that there's a logical reason for this disdain, however, if it can be shown that gold will (or would have) improve your investment performance. That leads into your second comment.
NiceUnparticularMan wrote:If the argument for doing so depends on an unreliable back-test, then I am just going to stick with that as my answer.
Like everyone else, I have no idea what the price of gold will be in the next forty years, but I do know what it has done for the last 40 years. Unless we were restricting our analysis to those countries whose currency was not connected to gold, I find any analysis of gold prices prior to that highly unreliable. But 40 years of data is nothing to sneeze at; that's longer than most of us will be in the accumulation phase at the least. And those data are quite clear; during the last 40 years, a small holding of gold (up to ~20% it seems to me) in an otherwise 'typical' portfolio with stocks and bonds would have reduced the portfolio's volatility without a corresponding reduction in returns such that it the safe withdrawal rate would be significantly higher than a portfolio comprised of only stocks and bonds in any ratio.

Now one could argue about whether that will continue to be the case going forward, but there's no disputing that it was true for the last 40 years.
I agree entirely with this. Historically, gold is a portfolio stabilizer, particularly in a dis-accumulation portfolio.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: an in depth look at gold

Post by NiceUnparticularMan »

willthrill81 wrote:I think you've hit the nail on the head as to why so many Bogleheads are opposed to gold. It has everything to do with the fact that it's an inanimate object. I don't think that there's a logical reason for this disdain, however, if it can be shown that gold will (or would have) improve your investment performance. That leads into your second comment.
"Would have" doesn't mean anything to me per se, because I can't go back and invest in the past. So I am going to set the requirement at "will."
Like everyone else, I have no idea what the price of gold will be in the next forty years, but I do know what it has done for the last 40 years. Unless we were restricting our analysis to those countries whose currency was not connected to gold, I find any analysis of gold prices prior to that highly unreliable. But 40 years of data is nothing to sneeze at; that's longer than most of us will be in the accumulation phase at the least. And those data are quite clear; during the last 40 years, a small holding of gold (up to ~20% it seems to me) in an otherwise 'typical' portfolio with stocks and bonds would have reduced the portfolio's volatility without a corresponding reduction in returns such that it the safe withdrawal rate would be significantly higher than a portfolio comprised of only stocks and bonds in any ratio.
If during the historic period of time in question there was what I called an "evolution" or a "structural break", or if during the future period you are interested in there will be such an evolution or structural break, then it doesn't matter how many years of data you are looking at--all of the data prior to that evolution or structural break is no longer going to be reliable in terms of predicting the future past that point.

In this case, we actually know for a fact there was an evolution/structural break during the historic period in question. So this is an easy one for me. In other cases we may not know for sure what has happened so far, but it is such an omnipresent danger, particularly when contemplating the future as well, that in my view it is basically impossible for a back-test to show conclusively what portfolio mix WILL be most efficient (or provide the highest return, lowest volatility, be safest, or so on) going forward. So, my requirement has not been met

As I mentioned before, back-testing is a weak tool in general. And in this case, we know at least one very good reason why it isn't reliable in this case, given the proposed historic period in question.
Now one could argue about whether that will continue to be the case going forward, but there's no disputing that it was true for the last 40 years.
Unfortunately for me, as an investor stuck in the present, that first part is the question that actually matters.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: an in depth look at gold

Post by willthrill81 »

NiceUnparticularMan wrote:Unfortunately for me, as an investor stuck in the present, that first part is the question that actually matters.
Well if you don't believe in historical results, I suppose you're left with intuition and/or blind luck.

The future will certainly look different than the past, but as Mark Twain says, it rhymes.
The Sensible Steward
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: an in depth look at gold

Post by NiceUnparticularMan »

willthrill81 wrote:Well if you don't believe in historical results, I suppose you're left with intuition and/or blind luck.
I think you can do a little better than that. For example, you can look at the kinds of assets available to you, and specifically what you are buying when you purchase them, and make some solid theoretical decisions about what it makes sense to buy in light of your goals. Diversification is also generally a good idea for pretty solid theoretical reasons. Within reason, seeking lower fees and tax efficiency are pretty good ideas. And so on.

But yes, in 20, 30, or 40 years, people will look back and figure out what would have been the ideal portfolio, and your odds of nailing it are dismal. Very likely you will have to settle for "could have been a lot better, but not too bad." And I have prepared myself for living with merely "not too bad".
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: an in depth look at gold

Post by packer16 »

Another data point is 200 year record for gold in Jeremy Siegel's "The Future for Investors", p. 171. Gold compounded in USD about 0.2% per year over 200 years & t-bill about 2.9% over the same period. If you are holding gold as hedge against volatility in the withdraw phase then hold a little at critical times of withdraw (like a few years before and after retirement). I look on gold and all commodities that are not consumed & easily re-generated (all except O&G) as volatility hedges or insurance. Just like insurance it costs you expected return so you should only buy it when you need it.

Packer
Buy cheap and something good might happen
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: an in depth look at gold

Post by willthrill81 »

packer16 wrote:Another data point is 200 year record for gold in Jeremy Siegel's "The Future for Investors", p. 171. Gold compounded in USD about 0.2% per year over 200 years & t-bill about 2.9% over the same period. If you are holding gold as hedge against volatility in the withdraw phase then hold a little at critical times of withdraw (like a few years before and after retirement). I look on gold and all commodities that are not consumed & easily re-generated (all except O&G) as volatility hedges or insurance. Just like insurance it costs you expected return so you should only buy it when you need it.

Packer
Very well said. I have my doubts that even most on this forum are truly aware of how detrimental volatility is to your portfolio in the withdrawal phase. Whether it's gold, commodities, or something else, reducing that volatility even at the expense of some returns is well worth it.

I really wonder if the 'seminal' studies on the topic of SWRs had taken this perspective whether we would be talking about the 5% or even 6% rule.
The Sensible Steward
Valuethinker
Posts: 49027
Joined: Fri May 11, 2007 11:07 am

Re: an in depth look at gold

Post by Valuethinker »

Commodore wrote:One thing that prevents me from investing in gold is technology. Being in the commodities business, we have seen how massive technological breakthroughs when extracting fossil fuels (Oil & Gas) has caused the overall pricing complex to be completely re-set to lower production costs and ultimately lower commodity prices. If this happened in gold mining, gold prices would in turn re-set to much lower levels. This is a huge downside risk IMO.
Swedroe mentions that Erb & Campbell note this.

i.e. that a technological change that would allow us to extract gold from, say, sea water, economically would more or less obliterate the value of gold*. Asteroid mining comes to mind as well.

In James Blish's Cities in Flight, people live essentially forever and the world's cities use "spin-dizzies" to fly around the Galaxy. But they are mostly broke- -they call themselves Okies. And so New York, to escape marauding space pirates, crosses a rift in the Galaxy and is out of touch for years. But they find huge germanium resources, and germanium is the de facto universal currency of galactic civilization.

They find their way back to civilized space, expecting to be rich. Instead, they discover that germanium is just a metal, and the currency is now the life extension drugs. They are, once again, a poor city begging for a living on the interstellar trade routes.

There is a Twilight Zone episode about bank thieves who arrange to have themselves put in suspended animation for 100 years so they can escape with their loot. They do so, but the ironic twist is the stolen loot is now worthless in the world of 2050 or whatever-- I forget as to why.

So, germanium my friends. Germanium.



* I am thinking we are talking controlled nuclear fusion to have energy so cheaply. But then, of course, transmutation of metals might be economic as well ;-).
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: an in depth look at gold

Post by lack_ey »

larryswedroe wrote:If gold is an inflation hedge then current price relative to inflation index should be a good indicator of it's value, again assuming that's it's purpose. That seems like a simple and logic concept since it is in fact why many buy it.
Larry
I don't know why they and you insist on calling it a price relative to the inflation index here in this context, as if you're looking at a price relative to a fundamental. It's just a real price (inflation-adjusted in USD terms) time series. There's no fundamental or flows-based explanation of returns, so I think the paper's formulation is unhelpful there.

There is just a hypothesis—supported in part by somewhat anecdotal long-term data—that the real price probably shouldn't vastly change over long periods of time. Also it's noted that within a span of 40 years or so, there's been almost a 6x span of real gold price.

Now, if the real price indeed should stay about constant over time, if this period studied gives us an unbiased view of the underlying behavior and is indicative of future behavior, then the average real price over this period may be a reasonable estimate of the average real price we'll see in the future (a period not starting tomorrow, but far enough in the future such that the current price doesn't have much residual influence... remember that obviously, prices are correlated in time). That's not a particularly strong premise, given a lot of the things that could change. Some have been mentioned above so I won't repeat what others have written. Furthermore, to show mean reversion, you have to do a lot more than demonstrate that price predicts price in the way given in the paper. You could do the same analysis on a memoryless random walk of prices that we pull out of a simulation and find that relationship.

Did you know that if you draw a line in hindsight about halfway through the middle of some squiggles, the squiggles will be above the line about half the time and below about half the time, and especially low values compared to the line will predict higher future returns, and especially high values compared to the line will predict lower future returns? Give me a break.

Even accepting all of the above, keep in mind that a ~40 year observation of a volatile asset will still result in a good deal of noise or error in our observation of what the average real price "should" be. There are a very wide ranges of sequences of real prices that would be consistent with all of the observations made and every explanation given in the papers or this thread. Each of these might well have happened instead, and maybe we would instead calculate a golden constant of 2 instead of 3.5. Or perhaps 9. Maybe we'd be sitting on a real price 60% below what it actually is, or perhaps 200% more. Statistically all these observations are fairly fragile, not particularly all that meaningful. Even if we assume a lot of things and agree on let's say a golden constant of 3.6 being the best estimate possible, it's probably not a very good estimate.

I couldn't necessarily tell you a better estimate, but I think any certainty, knowledge, information about the underlying is not as well grounded as some think. Gold is certainly diversifying, but if anybody (as some have here) wants to say for sure that it's diversifying in a good way, you need stronger evidence than anything presented here.
User avatar
CULater
Posts: 2832
Joined: Sun Nov 13, 2016 9:59 am
Location: Hic sunt dracones

Re: an in depth look at gold

Post by CULater »

I put 5% in a gold ETF because I have as much overvalued stock as I'm comfortable with, way much more in nil return bonds than I'm comfortable with, didn't want to invest more in either, and thought that gold might be a little more interesting than cash. When compared to krap, gold doesn't look so bad even if it's krap too. It's good to diversify your krap.
On the internet, nobody knows you're a dog.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: an in depth look at gold

Post by willthrill81 »

Valuethinker wrote:
Commodore wrote:One thing that prevents me from investing in gold is technology. Being in the commodities business, we have seen how massive technological breakthroughs when extracting fossil fuels (Oil & Gas) has caused the overall pricing complex to be completely re-set to lower production costs and ultimately lower commodity prices. If this happened in gold mining, gold prices would in turn re-set to much lower levels. This is a huge downside risk IMO.
Swedroe mentions that Erb & Campbell note this.

i.e. that a technological change that would allow us to extract gold from, say, sea water, economically would more or less obliterate the value of gold*. Asteroid mining comes to mind as well.

In James Blish's Cities in Flight, people live essentially forever and the world's cities use "spin-dizzies" to fly around the Galaxy. But they are mostly broke- -they call themselves Okies. And so New York, to escape marauding space pirates, crosses a rift in the Galaxy and is out of touch for years. But they find huge germanium resources, and germanium is the de facto universal currency of galactic civilization.

They find their way back to civilized space, expecting to be rich. Instead, they discover that germanium is just a metal, and the currency is now the life extension drugs. They are, once again, a poor city begging for a living on the interstellar trade routes.

There is a Twilight Zone episode about bank thieves who arrange to have themselves put in suspended animation for 100 years so they can escape with their loot. They do so, but the ironic twist is the stolen loot is now worthless in the world of 2050 or whatever-- I forget as to why.

So, germanium my friends. Germanium.



* I am thinking we are talking controlled nuclear fusion to have energy so cheaply. But then, of course, transmutation of metals might be economic as well ;-).
I think that we're a ways off from transmutation of metals. :wink:

We should have had Mr. Fusion by now too, but I guess the would-be creators didn't watch 80s movies. And frankly, if that kind of stuff actually transpires, it could lead to something resembling a Star Trek utopia. Think about what would happen to global economies if energy became essentially free and unlimited.

I'm not banking on any of it though.
The Sensible Steward
boglerdude
Posts: 928
Joined: Fri Jul 10, 2015 1:28 am

Re: an in depth look at gold

Post by boglerdude »

There will be increased mining for battery metals.

Is there a lot of gold as a by product?

edit: In other words, a way to get exposure to gold price while also investing in a productive mining company.
User avatar
raven15
Posts: 476
Joined: Sun Nov 30, 2014 7:01 pm

Re: an in depth look at gold

Post by raven15 »

boglerdude wrote:There will be increased mining for battery metals.

Is there a lot of gold as a by product?

edit: In other words, a way to get exposure to gold price while also investing in a productive mining company.
As far as I know gold is only a by product of mining chemically similar minerals such as silver and copper, which are likewise byproducts of mining gold. Lithium, for example, is completely different. Gold is too rare to try and deliberately extract it from any rocks that don't have a very high concentration (relatively speaking), so nobody would say "gee we already dug up these rocks, lets get extract gold from them while we're at it!" The scale of global mining probably won't change much as a result of more demand for batteries, it will just be a different type of mining.
It's Time. Adding Interest.
nimo956
Posts: 914
Joined: Mon Feb 15, 2010 5:07 pm

Re: an in depth look at gold

Post by nimo956 »

IngognitoUSA wrote:Physical gold as investment in not very good. However, if you had to flee Syria the only thing worthwhile you can carry out would be gold. Your property, stock investment and currency are probably worthless.

If your currency got demonetized as in India recently, gold would have been a good alternative to keep.
Yes, I imagine that we in the US are lucky that the dollar is considered the world's most dominant reserve currency. I'd think people in developing nations, whose governments' actions might undermine the value/stability of their currency, have a much different perspective. Even Iceland, which is a first-world country, nearly saw its currency collapse in 2008.
50% VTI / 50% VXUS
AlohaJoe
Posts: 6609
Joined: Mon Nov 26, 2007 1:00 pm
Location: Saigon, Vietnam

Re: an in depth look at gold

Post by AlohaJoe »

willthrill81 wrote:I think that we're a ways off from transmutation of metals. :wink:
(This is more an interesting aside rather than anything that I think will actually affect investing of gold or anything else in our lifetimes...)

NASA has an "Asteroid Redirect Mission" (that maybe was recently delayed/scrapped/something) that was going to bring several tons from an asteroid closer to earth for science/mining/whatever.

One possible target is Asteroid 16 Psyche which, by some estimates, has more gold than has been mined in thousands of years of human history.

So...I'm calling a bear market in gold starting in 2030! :twisted:
Post Reply